The marRAB multiple antibiotic resistance operon of Escherichia coli controls the expression of a large number of genes resulting in low level antibiotic and superoxide resistance through a complex network of reactions. MarR auto-represses the mar operon but is inactivated upon interaction with salicylate, losing its DNA binding capacity. This, in turn, results in derepression of the operon and expression of MarA, which activates the transcription of some 40 to 60 promoters (the mar/sox/rob regulon) including the marRAB promoter itself (auto-activation). We have focused recently on the capacity of MarA to activate these disparate promoters in an attempt to develop a mathematical model to describe the network. To do so we have constructed a large array of tester strains, each containing one of the promoters of the regulon linked to the lacZ gene. The strains also contain a high copy number plasmid whose expression of MarA (or the closely related transcriptional activator, SoxS) can be precisely controlled by the addition of IPTG to the medium (since the strains also carry an episome with the lacIq gene). Thus, we can assess in vivo transcriptional activation by measuring the equilibrium expression of beta-galactosidase as a function of IPTG concentration added to the medium. We find that the concentrations of IPTG required for half-maximum activation of the different promoters varies by greater than 50-fold. Thus it is possible to fully activate certain of the genes of the regulon at levels of MarA that have virtually no effect on other genes, a phenomenon we are tentatively referring to as 'gradational response'. A majority of the promoters were only activated at high levels of IPTG raising the possibility that these promoters were still not fully stimulated, but rather were limited by the levels of MarA obtainable. This, because MarA is rapidly degraded in vivo. The analyses were therefore repeated in strains defective for the lon protease, the principle protein responsible for MarA degradation. From these analyses we have determined the concentrations of IPTG for half-maximal activation and estimates of maximal activity for 14 of the promoters. To translate the IPTG concentrations into in vivo concentrations of MarA we have carried out extensive quantitative analyses by Western blots. Our collaborator, Michael Wall, is currently constructing a mathematical model consistent with the data. [unreadable] We and others have previously proposed models for the interaction of MarA with RNA polymerase requisite for transcriptional activation. The above analyses have thrown into doubt some aspects of those models. Consequently we are currently engaged in an extensive analysis of MarA mutants employing the above technique.[unreadable] This work was carried out in collaboration principally with Drs. J.L. Rosner and Michael Wall (Computer and Computational Sciences & Bioscience, Los Alamos National Laboratory, Mail Stop B256,Los Alamos NM 87545 USA).